Automatic Detection of Fake News

By Verónica Pérez-Rosas, Bennett Kleinberg, Alexandra Lefevre, Rada Mihalcea. 2017

936 citations recorded by the source corpus.

The proliferation of misleading information in everyday access media outlets such as social media feeds, news blogs, and online newspapers have made it challenging to identify trustworthy news sources, thus increasing the need for computational tools able to provide insights into the reliability of online content. In this paper, we focus on the automatic identification of fake content in online news. Our contribution is twofold. First, we introduce two novel datasets for the task of fake news detection, covering seven different news domains. We describe the collection, annotation, and validation process in detail and present several exploratory analyses on the identification of linguistic differences in fake and legitimate news content. Second, we conduct a set of learning experiments to build accurate fake news detectors, and show that we can achieve accuracies of up to 76%. In addition, we provide comparative analyses of the automatic and manual identification of fake news.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Sociology and Political Science

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Defending Against Neural Fake News (2019, background): “Efforts to automate fake news detection generally point out stylistic biases that exist in the text (Rashkin et al., 2017; Wang, 2017; Pérez-Rosas et al., 2018).”
Improving Wikipedia verifiability with AI (2022, background): “While previous works (Bowman et al., 2015; Wang et al., 2018; Camburu et al., 2018; Nie et al., 2019; Pérez-Rosas et al., 2017; Thorne et al., 2018; Thorne & Vlachos, 2018) have shown the ability of large neural networks to perform well on natural language understanding tasks, these results were mos”
Predicting Factuality of Reporting and Bias of News Media Sources (2018, background): “, 2018) and (Pérez-Rosas et al., 2018), who use news articles whose factuality has been manually checked and annotated.”
FakeBERT: Fake news detection in social media with a BERT-based deep learning approach (2021, background): “One of the researches, Perez-Rosas et al [29] have explained the automatic identification of fake content in online news articles.”
Fake News Detection Using Machine Learning Ensemble Methods (2020, methodology): “In the current fake news corpus, there have been multiple instances where both supervised and unsupervised learning algorithms are used to classify text [20, 21].”
The Limitations of Stylometry for Detecting Machine-Generated Fake News (2019, neutral): “2011; Feng, Banerjee, and Choi 2012; Afroz, Brennan, and Greenstadt 2012), fake news (Rashkin et al. 2017; Pérez-Rosas et al. 2018), or other false or illegal content (Choshen et al.”
Mining Dual Emotion for Fake News Detection (2019, background): “, designing a classifier to judge a given news piece as real or fake) by leveraging signals from text [5, 32, 34], images [20, 35], or social contexts [17, 24, 26– 28, 39, 40].”
Identifying the sentiment styles of YouTube’s vloggers (2018, methodology): “natural language and its linguistic style and content are applied on a large scale. Applications range from the automated detection of fake reviews (Ott et al., 2013), fake news (Thorne et al., 2018; Pérez-Rosas et al., 2018) and lies and truths (Mihalcea and Strapparava, 2009; Kleinberg et al., 201”
Fake News Detection on Social Media using Geometric Deep Learning (2019, background): “Content-based approaches, which are used in the majority of works on fake news detection, rely on linguistic (lexical and syntactical) features that can capture deceptive cues or writing styles [1, 32, 30, 29, 28].”
Combating Fake News (2019, background): “[82] found that fake news articles contained more social words, with more verbs and temporal words, suggesting that the text tends to be focused on the present and future, instead of being more objective and factual.”
SemEval-2019 Task 8: Fact Checking in Community Question Answering Forums (2019, background): “The reason for this is the way the data news” and click-bait detection (Hardalov et al., 2016; Karadzhov et al., 2017a; Pan et al., 2018; Pérez-Rosas et al., 2018).”
False Information on Web and Social Media: A Survey (2018, methodology): “[74] shows the changing effectiveness of text-based features.”
Explainable Automated Fact-Checking: A Survey (2020, methodology): “, the choice between binary (Pérez-Rosas et al., 2018) and graded classification (Vlachos and Riedel, 2014; Wang, 2017) informs the definition of veracity.”
A Survey of Fake News (2020, methodology): “…further compares such frequency with that in other news articles in the corpus, such as In general, TF-IDF can be applied at various language levels, and so can n -gram models, which enable capturing the sequence of words (POS tags, rewrite rules, etc.) [Feng et al. 2012; Pérez-Rosas et al. 2017].”
Information Retrieval: A Biomedical and Health Perspective (2020, background): “One possible solution is to certify the research with a trusted third party [310].”
A Survey on Automated Fact-Checking (2021, background): “, 2017) Triple 18,431 KG 2 Classes Google/Wiki/WSDM En UFC (Kim and Choi, 2020) Triple 1,759 KG 2 Classes Wiki En LieDetect (Mihalcea and Strapparava, 2009) Passage 600 ✗ 2 Classes News En FakeNewsAMT (Pérez-Rosas et al., 2018) Passage 680 ✗ 2 Classes News En FEVER (Thorne et al.”